Making Better Drugs for Children with Cancer

EXECUTIVE SUMMARY

The oncology world is justifiably proud of the fact that cancer during childhood and adolescence has been transformed from a death sentence into long-term survival for most of those affected, at least in the United States and other countries with access to good health care. Today, major improvements in survival from childhood cancers that still have a poor prognosis—and reducing the substantial short- and long-term adverse effects of current therapies—will come only with new treatments. The near absence of research in pediatric cancer drug discovery threatens to halt the progress in childhood cancer treatment achieved during the past four decades.

The achievements of the past 40 years are testament to the efforts of pediatric oncologists, radiation oncologists, and surgeons nationally and internationally who systematically evaluated the anticancer drugs developed during this period, largely for cancers in adults, for use in children. They built up effective regimens primarily by increasing the intensity of therapy in high-risk patients. Even now, adjustments in treatment regimens are continually being tested to enhance effectiveness and reduce unwanted effects, but further improvements through intensification of therapy with the same drugs are likely to be small in magnitude.

As devastating as cancer is among children in the United States (cancer remains the leading cause of death by disease in U.S. children between 1 and 15 years of age), the number of children affected is small. From the point of view of companies developing drugs and other agents to treat



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
Making Better Drugs for Children with Cancer Making Better Drugs for Children with Cancer EXECUTIVE SUMMARY The oncology world is justifiably proud of the fact that cancer during childhood and adolescence has been transformed from a death sentence into long-term survival for most of those affected, at least in the United States and other countries with access to good health care. Today, major improvements in survival from childhood cancers that still have a poor prognosis—and reducing the substantial short- and long-term adverse effects of current therapies—will come only with new treatments. The near absence of research in pediatric cancer drug discovery threatens to halt the progress in childhood cancer treatment achieved during the past four decades. The achievements of the past 40 years are testament to the efforts of pediatric oncologists, radiation oncologists, and surgeons nationally and internationally who systematically evaluated the anticancer drugs developed during this period, largely for cancers in adults, for use in children. They built up effective regimens primarily by increasing the intensity of therapy in high-risk patients. Even now, adjustments in treatment regimens are continually being tested to enhance effectiveness and reduce unwanted effects, but further improvements through intensification of therapy with the same drugs are likely to be small in magnitude. As devastating as cancer is among children in the United States (cancer remains the leading cause of death by disease in U.S. children between 1 and 15 years of age), the number of children affected is small. From the point of view of companies developing drugs and other agents to treat

OCR for page 1
Making Better Drugs for Children with Cancer cancer, the pediatric cancer drug market is often well below the radar screen, and typically it has not made business sense to invest in research and development for these cancers. Many drugs developed for adults have been found effective in children, in large part because most have a generalized affinity for cancer cells. The dark side of this characteristic is that noncancerous tissue may be damaged in the process, causing the well-known adverse effects of anticancer drugs. But in fact, the biological and clinical characteristics of nearly all childhood cancers differ substantially from adult cancers. Over the past few years, differences at the molecular level have been documented for all the major childhood cancers, and herein lies the promise: the molecular abnormalities represent a place to start searching for drug “targets.” Cutting-edge science notwithstanding, market forces are not sufficient to drive the process and bring to the bedside new drugs for children with cancer. Because so much of the technical capacity for drug discovery and development for pediatric cancers already exists—much of it supported by the National Institutes of Health (NIH) intramurally or extramurally, as well as in the private for-profit sector—it is possible that these drugs could emerge from an alternate pathway. Specifically, a “public– private partnership” could knit together the pieces in a virtual research and development (R&D) network. Networks such as this are relatively new, but are working well for cystic fibrosis, tuberculosis, malaria, and other neglected tropical diseases. The resources already in place for pediatric cancers are poised for this development. Recommendation 1: A new public–private partnership, involving government, industry, academic and other research institutions, advocacy groups, philanthropies, and others, should be formed to lead pediatric cancer drug discovery and development. R&D specifically for childhood cancers is not the only way to improve treatment. New drugs in development for adult cancers may prove useful for children, including some that, in the final analysis, fail to provide significant benefit to adults. This poses a dilemma for companies that would never be able to recoup development costs if a full-scale effort resulted only in a drug for children. At whatever stage a product is in development, if this is the case, the government should consider taking over the development process, either directly or through funding external work. Recommendation 2: The National Cancer Institute should assume responsibility as the developer of last resort for agents that show promise only in children if companies decide not to proceed with full-scale development. Even when an agent is shown effective in adults, historically, long delays have intervened between the time testing is begun in adults and in children. Because there are relatively few children with these cancers, clini-

OCR for page 1
Making Better Drugs for Children with Cancer cal trials can extend years longer than they do in adults. There are ways to shorten this period, mainly by beginning trials earlier. The National Cancer Institute (NCI) has begun to facilitate this by funding the first consolidated preclinical testing program for pediatric cancers, so that the necessary steps before the first use in children are completed as quickly as possible. Other steps must be taken to encourage companies to allow drugs to enter pediatric clinical trials earlier. Legislative initiatives over the past decade to encourage broader testing of all types of drugs in children could play a role in this, although because of issues specific to cancer, they have as yet had little impact on testing or labeling cancer drugs for children. Recommendation 3: The pharmaceutical industry, National Cancer Institute, and Food and Drug Administration should act to reduce the delay in beginning pediatric clinical studies of agents in development for adult cancers.

OCR for page 1
Making Better Drugs for Children with Cancer INTRODUCTION The successes that have been achieved in treating childhood cancers stand as beacons against the less dramatic improvements for adults with cancer. Progress began to accelerate in the 1960s and 1970s, as treatment regimens were built up, primarily by building combinations of chemotherapeutic drugs. But since that time, progress in most childhood cancers has come mainly from increasing the dose intensity of existing drugs in children at increased risk of disease recurrence, better understanding of the diseases, and infrequently from adopting new drugs. Acute lymphocytic leukemia (ALL), the most common childhood cancer, is a case in point: developments in the laboratory—discoveries about the biologic underpinnings of leukemia—have led to more precise subcategorization and risk stratification of patients. Better understanding of how to use the existing drugs in light of the biologic information was more important than the development of new drugs (including etoposide; introduced in 1983 and the last major addition to ALL regimens, it is featured only in treating a small minority of high-risk patients). Nonetheless, in the past 20 years, 5-year survival has improved from 50 percent to 80 percent (U.S. Cancer Statistics Working Group, 2004). Nearly all drugs currently used to treat children with cancer, with the exception of a few antileukemia drugs developed in the 1950s and early 1960s, were developed for adult cancers and then found effective in children. The remarkable success of regimens built with these drugs might argue for continuing on this path. But it is also true that most common types of childhood cancer are distinct from adult cancers clinically, pathologically, cytologically, and in the molecular abnormalities that underlie them (Table 1). If the discovery and development of new agents for childhood cancers were to capitalize on today’s science—which in large measure it does not—there is every reason to believe that cure rates could be improved for all pediatric cancers, including those for which current long-term survival is very low, such as brain tumors. Moreover, advances based on more targeted agents could entail considerably fewer short- and long-term toxicities, which themselves become lifelong medical problems for many survivors (IOM, 2003). This report identifies the major issues to be addressed in developing new agents for childhood cancers, the gaps in research and development, and the steps that have been suggested to move the process forward, and makes a new proposal to capitalize on today’s science to bring new treatments to children’s cancers. The Institute of Medicine (IOM) has previously issued reports related to the development and testing of drugs of all types for children (IOM, 2000) and the ethical conduct of clinical research in-

OCR for page 1
Making Better Drugs for Children with Cancer volving children (IOM, 2004), both of which include discussions relevant to the issues explored here. A key point is that the financial market for childhood cancer drugs is very small—below the dollar value needed to interest profit-making drug and biotechnology companies. Even incentives such as orphan drug provisions and the recent “pediatric incentive” offered for testing products in children cannot tip the balance sufficiently. For-profit companies are unlikely to invest in the early discovery stages of R&D for drugs that, even if successful, have such a limited market (assuming the drug is useful only for children). At least in some cases, however, companies may find it worth-while to engage in late-stage development and manufacturing of low-volume drugs once their potential has been established (i.e., “de-risked” from a development perspective). CHILDHOOD CANCERS: BACKGROUND Childhood cancer is rare. In the United States, an estimated 11,900 children and adolescents under age 20 were diagnosed with cancer in the year 2001 and about 2,200 died (Figures 1 and 2). These cases represent an incidence rate of about 16 per 100,000 per year (roughly 1 per 6,400 children per year). Between birth and 20 years of age, about 1 in 333 Americans develops cancer. A few types of cancer predominate—leukemias, and tumors of the brain and nervous system, the lymphatic system, kidneys, bones, and muscles. For the most part, these cancers are distinct biologically and clinically from the cancers common in adults, even if the names are similar. Although most children and adolescents now survive their cancer, it is the third leading cause of death among children ages 1 to 4, and second only to accidents among children ages 5 to 14 (Minino and Smith, 2001) (Table 2). Genetic factors and certain prenatal exposures (e.g., ionizing radiation, diethylstilbestrol [DES]) and postnatal exposures (e.g., viruses, ionizing radiation) are known to increase the risk of developing some childhood cancers, but for most cases, the cause remains unknown, precluding strategies for prevention. Cancer in the first half of the 20th century was nearly synonymous with death for children. Fewer than 10 percent survived. But beginning in the 1960s, discoveries of effective treatments, particularly for the leukemias, began to change that picture. Improvements since the 1960s have been steady and significant, dramatic in some cases. For some types, however—particularly brain tumors, acute nonlymphocytic leukemias, and metastatic solid tumors—treatment and survival are still poor (Table 3). (See Appendix A for a review of the epidemiology and treatment of childhood cancers.)

OCR for page 1
Making Better Drugs for Children with Cancer TABLE 1 Molecular Abnormalities That Are Common in Childhood Cancer but Rare or Nonexistent in Adult Cancer Disease Translocation Fusion/Molecular Abnormality New Patients per Year in U.S. Children (0–19 y)* Alveolar rhabdomyosarcoma t(2;13)(q35;q14) t(1;13)(p36;q14) PAX3-FKHR PAX7-FKHR <200 Ewing’ s sarcoma family tumor t(11;22)(q24;q12) t(21;22)(q22;q12), t(2;11;22), t(12;22) t(7;22)(p22;q12) t(17;27)(q12;q12) t(2;21;22) EWS-FLI1 EWS-ERG EWS-ETV1 EWS-E1AF EWS-FEV <400 Malignant melanoma of soft parts t(12;22)(q13;q12) EWS-ATF1   Desmoplastic small round cell tumor t(11;22)(p13;q12) EWS-WT1   Extraskeletal myxoid chondrosarcoma t(9;22)(q22;q12) t(9;17)(q22;q11.2) EWS-CHN (TEC) RBP56 (hTAFII68)-CHN(TEC)   Synovial cell sarcoma t(X;18)(p11;q11) SYT-SSX1 or SSX2   Myxoid liposarcoma t(12;16)(q13;p11) t(12;22)(q13;q12) t(12;22;20)(q13;q12;q11) TLS(FUS)-CHOP EWS-CHOP EWS-CHOP   Congenital fibrosarcoma t(12;15)(p13;q25) ETV6-NTRK3  

OCR for page 1
Making Better Drugs for Children with Cancer Dermatofibrosarcoma t(17;22)(q22;q13) COL1A1-PDGFB   Anaplastic large cell lymphoma t(2;5) NPM-ALK <150 Acute megakaryoblastic leukemia in infants t(1;22)(p13;q13) OTT-MAL (RBM15-MLK1)   B-lineage ALL t(12;21)(p13;q22) TEL-AML1 AML-ETO MLL (HOXA9, Meis1) <2,500 Juvenile myelomonocytic leukemia   Somatic PTPN11 mutations NF1 GM-CSF hypersensitivity Overactivation of SHP   Other molecular lesions in pediatric cancers Retinoblastoma Rb gene (13q14) deletion (somatic or germ line mutation) <160 Neuroblastoma N-Myc amplification <370 Wilms’ tumor WT1 deletion <300 Malignant rhabdoid tumor INI1 deletion   Medulloblastoma Abnormalities in Ptc gene pathway ErbB2 overexpression <450 Glial tumors Arf/hdm/p53, ALK? <900 Novel surface antigens as putative targets: CD19 in B-lineage ALL; CD7 in T-ALL; NG2 in MLLr; stage-restricted antigens in embryonal tumors SOURCE: Information compiled by W. Evans, J. Downing, T. Golub, S. Morris, M. Smith, and C-H Pui. December 2002. *Estimated number of cases with the particular molecular subtype, based on data from year 2000.

OCR for page 1
Making Better Drugs for Children with Cancer FIGURE 1 Childhood cancer cases diagnosed in the United States, 2001, Ages 0–19. SOURCE: U.S. Cancer Statistics Working Group (2004). FIGURE 2 Childhood cancer deaths in the United States, 2001, Ages 0–19. SOURCE: U.S. Cancer Statistics Working Group (2004).

OCR for page 1
Making Better Drugs for Children with Cancer AGENTS IN USE AND IN TESTING FOR CHILDREN WITH CANCER Most of the drugs and biologics used routinely in primary treatment regimens for pediatric cancers were approved before 1990, about half before the mid-1980s (Table 4). As for adults with cancer, treatment with multiple combinations of drugs over an extended period is the norm in children (see Appendix A). The drugs used are all primarily used to treat cancer in adults and were initially approved for that purpose. Some either have been subsequently approved by the Food and Drug Administration (FDA) for pediatric indications or provide some pediatric dosing information in the label. For the first time in more than a decade, initial approval of a new cancer drug was granted for a pediatric indication: clofarabine to treat refractory or relapsed acute lymphocytic leukemia in children was approved by the FDA in late December 2004 (U.S. Food and Drug Administration, 2004). Clofarabine has orphan status,1 which provides for 7 years of exclusive marketing once the drug is approved, and the FDA has granted an additional 6 months of exclusivity under the pediatric incentive provision of the Best Pharmaceuticals for Children Act (discussed later). An early clinical trial (Phase I) in adult patients with advanced solid tumors has been completed, but applications for adult indications are likely years away. Medicines in Development for Children 2004 (PhRMA, 2004), a publication of the Pharmaceutical Research and Manufacturers of America (PhRMA), lists 32 products being tested for childhood cancers. These include drugs and biologics, about half of which are already approved for adult indications. Of those not yet approved, all are in more advanced testing in adults than in children. For 4 of the 32 agents, applications for pediatric labeling have been submitted, 2 of them are in Phase III trials, and another 3 are in multiple-phase trials that include Phase III. The others are in Phase I and II trials. THE EXISTING CAPACITY FOR DEVELOPING NEW AGENTS SPECIFICALLY FOR CHILDHOOD CANCERS Considerable basic research on childhood cancers has resulted in the identification of molecular abnormalities unique to those cancers (Table 1). If pursued, some of these could lead to new treatments, as has been the case 1   Orphan status is a designation under the Orphan Drug Act for a medication used to treat diseases that occur rarely (less than 200,000 prevalent cases in the population), so there is little financial incentive for industry to develop drugs to treat those diseases. Orphan drug sponsors are eligible for tax credits and exclusive marketing rights for the product.

OCR for page 1
Making Better Drugs for Children with Cancer TABLE 2 Number of Deaths and Death Rates for the 10 Leading Causes of Death in Specified Age Groups, United States, 2000   Ages 1–4 Years Cause of Deatha Rankb Number Rate All causes — 4,942 32.6 Accidents 1 1,780 11.7 Motor vehicle accidents   630 4.2 All other accidents   1,150 7.6 Congenital malformations, deformations, and chromosomal abnormalities 2 471 3.1 MALIGNANT NEOPLASMS 3 393 2.6 Assault (homicide) 4 318 2.1 Diseases of the heart 5 169 1.1 Influenza and pneumonia 6 96 0.6 Septicemia 7 91 0.6 Certain conditions originating in the perinatal period 8 84 0.6 In situ neoplasms, benign neoplasms, and neoplasms of uncertain or unknown behavior 9 56 0.4 Cerebrovascular diseases 10 45 0.3 Chronic lower respiratory diseases — — — Human immunodeficiency virus (HIV) disease — — — Intentional self-harm — — — All other causes — 1,439 9.5 NOTE: Data are based on continuous file of records received from the states. Rates per 100,000 population in specified group. Figures are based on weighted data rounded to the nearest individual, so categories may not add to totals. for targeted agents for adult cancers. However, there are no major R&D programs in either industry or the government devoted to developing new drugs for childhood cancers. As already explained, even with incentives such as orphan drug provisions, the market for pediatric cancer drugs is too small for pharmaceutical companies to recoup their investments in developing new products, and full-scale drug development is not considered a government function.

OCR for page 1
Making Better Drugs for Children with Cancer Ages 5–14 Years Ages 15–24 Years Rank Number Rate Rank Number Rate — 7,340 18.5 — 30,959 80.7 1 2,878 7.3 1 13,616 35.5   1,716 4.3   10,357 27.0   1,163 2.9   3,259 8.5 3 387 1.0 6 425 1.1 2 1,017 2.6 4 1,668 4.3 4 364 0.9 2 4,796 12.5 6 236 0.6 5 931 2.4 9 83 0.2 8 188 0.5 — — — — — — — — — — — — 8 106 0.3 — — — 10 78 0.2 7 193 0.5 7 130 0.3 9 180 0.5 — — — 10 178 0.5 5 297 0.7 3 3,877 10.1 — 1,764 4.4 — 4,907 12.8 aBased on the Tenth Revision, International Classification of Diseases, 1992. bRank based on number of deaths. SOURCE: Minino and Smith (2001). Even though no comprehensive R&D program exists to develop drugs for childhood cancers, the components that would constitute such a pipeline do exist—in universities, in academic medical centers, within pharmaceutical and biotechnology companies, and to a surprising degree, within NCI (and other parts of NIH). NCI’s role in supporting basic research is well known, but many of its activities in later stages of drug discovery and development are not.

OCR for page 1
Making Better Drugs for Children with Cancer relevant to treating children becomes available. First, there is the problem of finding enough eligible children to enroll in the trial (although for most pediatric cancers, this may not be not a significant limitation; the number of patients seeking new therapies exceeds the number of clinical trials that make such investigational therapies potentially available). Second, and more amenable to a solution, is that the lag between beginning adult trials and pediatric trials is longer than it has to be: there is strong interest on the part of the pediatric cancer research community in initiating the pediatric trials earlier than they have been. Irinotecan, a relatively new cytotoxic drug used primarily for colorectal cancer, is an example of a drug developed for adult malignancies that was tested against a variety of pediatric cancers in children who had not responded to other drugs, only after the drug was approved by the FDA for use in adults. The first Phase I trial results in adults were published in Japan in 1991, followed by Phase I trial results in the United States published in 1993. An accelerated approval was granted by the FDA in 1996, based on Phase II trial results in adults. It was only then, in 1996, that Phase I trials in children were initiated. The first were completed in 1998, when Phase II trials began for irinotecan, in combination with other agents for various solid and central nervous system tumors that had not responded to available treatments and untreated metastatic rhabdomyosarcoma. The drug was not deemed effective in these pediatric trials, which were completed in 2003, 10 years after publication of the first Phase I trial results in adults. Having carried out and reported the studies to the FDA, the sponsor fulfilled its agreement for pediatric information and will get an additional 6 months of exclusivity for irinotecan, under the pediatric incentive provision of the Best Pharmaceuticals for Children Act (discussed in more detail later). Because the drug was not effective in children in the doses and schedules studied, there will be no change in the product label. However, it still appears promising in certain pediatric cancers and continues to be studied (Personal communication, M.P. Link, Stanford University School of Medicine, January 25, 2005). Starting Pediatric Trials Earlier in the Drug Development Process It is important to understand why pediatric trials are not started earlier as a routine matter. Although other factors can be important, the key decisions on timing of trials are made by a drug’s developer, which is usually (especially by the time a drug is in clinical trials) a for-profit pharmaceutical or biotechnology company. This requires that the sponsor be willing to carry out the trials or allow others to do so. Even if the manufacturer (and patent holder) is not the trial sponsor (e.g., if NCI sponsors the trial), the fact that pediatric trials are being conducted could imply a com-

OCR for page 1
Making Better Drugs for Children with Cancer mitment to proceed with the drug development process for as long as results warrant. For a drug that has not yet proven effective for an adult indication (i.e., trials are in early stages), a hypothetical risk is that the drug will prove useful only for a pediatric indication, and not the adult indication for which it was actually being developed. A pediatric-only market would not be big enough to attract a large pharmaceutical company economically in investing in further development or maintaining a drug in production, in the absence of an astronomical per-course price. A second hypothetical concern for companies is the possibility that approval of a new drug in adults could be delayed if a serious, unexpected adverse event occurred in a pediatric trial. For these reasons many companies are reluctant to begin pediatric testing until effectiveness in adults is more certain, often only occurring following completion of adult Phase III trials. In a conservative situation, the pediatric trial may not commence until the company has obtained FDA approval for the new agent in adults. It should be appreciated that even adult cancer markets are relatively small compared to the market for drugs for cardiovascular conditions, for example. Despite the prevalence of the above concerns, there has been no documented instance of either a cancer drug being found effective exclusively for children, or an unexpected severe toxicity arising in a pediatric trial affecting approval for adults (although it should also be recognized that the opportunity for either concern to manifest itself has been very limited). The line of reasoning by companies described here, however, was affirmed generally by industry participants in a workshop held for this project (January 9, 2003, IOM Pediatric Oncology workshop). Even in the absence of data supporting the existence of such risk, companies may not see it in their interest to test earlier in children. Completing Pediatric Clinical Trials More Quickly There are limited ways to shorten the time it takes to conduct clinical trials to determine whether a drug is of benefit to patients. In the future, there may be validated surrogate end points that could cut the time to assessing clinical benefit, but these do not yet exist for pediatric cancers. Another way to improve timeliness is for patients to be entered over a shorter time span. The potential for this is also limited because a large proportion of children with cancer—about 5,000 per year—are already routinely enrolled in clinical trials, and enrollment in Phase III trials typically takes several years. Faster recruitment could be accomplished if there were fewer trials going on for each type of childhood cancer, or for rare pediatric tumors, if greater international collaboration could be arranged. Both of these avenues suggest a centralized process for deciding which new

OCR for page 1
Making Better Drugs for Children with Cancer agents should be given priority for clinical trials, and for organizing the trials. The latter exists through the Children’s Oncology Group, but the former has received less attention. The threshold for starting a drug through pediatric clinical trials is generally higher than it is for beginning trials in adult cancers for two main reasons (both of which result in few patients who could enroll in such trials). First is the relative rarity of cancer in children, and second is the relatively high success rate of first-line treatment for those with the most common forms of pediatric cancers. The price of this success—including the short- and long-term adverse effects—is to limit the number of patients eligible to enroll in a clinical trial of a new agent. If trials are to be completed in a timely manner, there can be only a limited number open for any type of cancer. Whether agents are prioritized deliberately among the pool of possibles, or less systematically (by decisions not to go ahead with a new trial because others are already ongoing, regardless of the promise of the agent), the development of new agents in the future ultimately will be “rationed” by the supply of patients. THE NEED TO PRIORITIZE AGENTS FOR TESTING IN CHILDREN WITH CANCER The case is made that, in absolute numbers, childhood cancer is rare (11,900 per year under age 20 in the United States). Even rarer are children for whom current treatments are ultimately unsuccessful—2,200 died each year at the beginning of the 21st century (U.S. Cancer Statistics Working Group, 2004). Compare this with the situation for adults: More than 200,000 women are diagnosed with breast cancer each year and a similar number of men are diagnosed with prostate cancer. Forty thousand women and 30,000 men die from these cancers each year. Similarly large numbers of people develop lung and colorectal cancer (American Cancer Society, 2004). (There are, of course, adult cancers that are as rare as childhood cancers, and the problems of developing and testing new agents for them may be equally daunting.) A rational approach to prioritizing drugs for testing in children with cancer ideally should depend on a consensus of all constituents in the pediatric oncology community (academic researchers, the FDA, NCI, companies, and advocates) that the best scientific data should guide which new agents go into the pediatric clinic. Of course, profit potential and marketing strategies drive many industry decisions about which new drugs are developed for adults, whether it is between cancer and other indications or within cancer. However, other variables unrelated to consensus about the science also play an influential role, including:

OCR for page 1
Making Better Drugs for Children with Cancer The FDA’s decision to issue a written request for pediatric studies under the provisions of the Best Pharmaceuticals for Children Act (BPCA); A company’s willingness to respond to the FDA’s written request for pediatric studies; Recommendations for trials by the pediatric oncologists in the Children’s Oncology Group and pediatric oncology therapeutic consortia; and Whether companies choose to continue to develop a drug for adults beyond Phase I. The current difficulties in determining which agents are most promising stem, in large part, from the absence of industry-conducted laboratory research on drugs in their development pipeline that focus on pediatric tumor targets. This may, in part, be remedied in the coming years as the pediatric preclinical testing program moves ahead. TESTING AND LABELING OF DRUGS FOR PEDIATRIC USE Relatively few pharmaceutical labels contain information about their use in children because most products have not been tested in children, or the information is only in the published medical literature. Although many drugs are used safely and effectively even in the absence of this information in the product label, serious adverse events that might have been prevented with direct testing in children have occurred at times. The well known examples do not involve cancer drugs, but the lack of labeling applies across all drug categories. Deaths caused by the antibiotic chloramphenicol (Powell and Nahata, 1982); pediatric kernicterus (a yellow staining in the brain that causes a form of cerebral palsy) from sulfa drugs; seizures and cardiac arrest caused by bupivacaine, a commonly used anesthetic; and hazardous interactions between erythromycin and midazolam, a sedative, all might have been avoided if the drugs had been tested in children before being prescribed widely.2 Drugs are nearly always approved by the FDA to treat disease in adults, with dosages and other labeling elements based on the studies conducted to gain approval. Once marketed, the drugs are prescribed for children in adjusted dosages at the discretion of the prescribing physician, based on experience, consultation with colleagues, published papers, and other information. Until recently, companies have had no legal requirement, and little 2   Specific Requirements on Content and Format of Labeling for Human Prescription Drugs; Revision of “Pediatric Use” Subsection In the Labeling, Part II. 21 CFR § 201 (1994).

OCR for page 1
Making Better Drugs for Children with Cancer incentive, to test drugs in children, with its attendant liabilities and operational complexities. But the medical community and the government perceived a need for better guidance on the use of products by children. Over the past decade, the FDA and Congress have taken two broad approaches to increasing the available information, referred to as (1) “the pediatric rule,” which has evolved into a federal law requiring that particular types of studies be performed in the course of the new drug approval process for drugs likely to be widely used in children, and (2) “the pediatric incentive,” which offers 6 months of added sales exclusivity for products already on the market that are still under patent or otherwise have market exclusivity (e.g., based on orphan drug status). The history of each approach is described in the following section. These initiatives have thus far had only a limited effect on drugs for childhood cancer, but are potentially important precedents that may play a more important role in the future, as certain issues are resolved. Where labeling is concerned, the issue that sets cancer drugs apart from drugs for other conditions is that the cancers for which they are used in children are different biologically and clinically from cancers affecting adults, as explained earlier in this report. Clinical experience in adults is of limited relevance to use in children, on whom the drugs will be used for different types of cancer, in regimens with different drugs than used in adults. But the decision to test a drug for cancer in children becomes part of the larger issue of prioritizing clinical trials in pediatric cancers, which is critical because of the small numbers of children eligible to participate. Testing one agent often preempts testing another one because of the small numbers of patients available for clinical trials. The Pediatric Rule The Short-Lived 1994 Pediatric Rule In 1994, the FDA assembled a list of the 10 most widely prescribed drugs in outpatient pediatrics (albuterol, Phenergan, ampicillin injections, Auralgan, Lotrisone, Prozac, Intal-aerosol, Zoloft, Ritalin, and Alupent). The approved labeling information for all 10 drugs—none of which is used to treat cancer—was inadequate or nonexistent for use in children. To remedy the situation, the FDA developed a “Final Pediatric Rule,” which it issued as a regulation in 1994.3 The rule “allowed efficacy data from adult 3   Specific Requirements on Content and Format of Labeling for Human Prescription Drugs; Revision of “Pediatric Use” Subsection In the Labeling, Part II. 21 CFR § 201 (1994).

OCR for page 1
Making Better Drugs for Children with Cancer studies to be extrapolated to a pediatric population, if the disease under study existed in both the pediatric and adult populations and the response to therapy was substantially the same”—a condition that does not hold for cancer. Drug manufacturers were required to examine the data they already had and determine whether the information was sufficient to support pediatric labeling information. Full-scale clinical trials in children were not required: Clinical data derived from trials in adults could be coupled with pediatric-specific pharmacokinetic and adverse reaction data. If existing data were insufficient, manufacturers were not required to conduct new studies. They only had to include this wording on the label: “Safety and effectiveness in pediatric patients have not been established.” Furthermore, the rule only applied to drugs that were already on the market. The 1994 Rule had little effect on drugs for childhood cancer. Many companies promised to conduct studies in the future, but it was clear that a voluntary approach was not substantially increasing the number of products with pediatric labeling information. Under the next version of the rule, testing was mandatory. The 1998 Pediatric Final Rule and the Pediatric Research Equity Act of 2003 The 1998 Pediatric Final Rule,4 which took effect in April 1999, “man-dates pediatric studies if an application for a claim is under review and the proposed indication is for a disease that exists in both adults and children.” This gave the FDA discretion to mandate specific pediatric studies and stated further that the “applicant also may be required to develop a pediatric formulation for a drug product that represents a meaningful therapeutic benefit to such patients over existing therapies.” Companies could request waivers if the mandated studies were highly impractical or if there was evidence that the product would be ineffective or unsafe for children. The 1998 Rule also had provisions for testing drugs already on the market, with the FDA bearing responsibility to demonstrate a need for testing. The 1998 Rule, similar to the earlier version, applies only to diseases that are the same in adults and children. This has limited its application to cancer drugs because the prevalent cancers of children and adults are different (Pervan et al., 2001). No labeling changes have yet been reported for cancer drugs as a result of the rule. In 1999, the Association of American Physicians and Surgeons, the 4   Regulations Requiring Manufacturers to Assess the Safety and Effectiveness of New Drugs and Biological Products in Pediatric Patients; Final Rule. 21 C.F.R. § 201, 312, 314, 601 (1998).

OCR for page 1
Making Better Drugs for Children with Cancer Competitive Enterprise Institute, and Consumer Alert filed a citizen petition with the FDA challenging the 1998 Rule. In 2000, the FDA denied the petition, and the same group brought suit against the FDA in U.S. District Court, asserting that the “Pediatric Rule exceeds the FDA’s statutory authority and that the Rule’s promulgation was arbitrary and capricious” (U.S. District Court for the District of Columbia, 2002). In October 2002, the District Court of the District of Columbia agreed, striking down the Pediatric Rule (U.S. District Court for the District of Columbia, 2002). The American Academy of Pediatrics and the Elizabeth Glaser Pediatric AIDS Foundation appealed (Pediatric AIDS Foundation, 2002), but the appeal was preempted by passage of the Pediatric Research Equity Act of 2003 (Pub. L. No. 108–155) in December 2003, enacting the provisions of the Pediatric Rule into law for 5 years. Pediatric Subcommittee of the FDA Oncologic Drugs Advisory Committee Since 2000, the FDA has held public meetings of a new Pediatric Subcommittee of the Oncologic Drugs Advisory Committee to discuss broad principles that would underlie the agency’s application of the Pediatric Rule to oncology drugs, including the question of equivalence of adult and pediatric diseases. Congress subsequently established this Subcommittee in law in the Best Pharmaceuticals for Children Act in January 2002, broadening membership to include representatives of NCI, patient advocates, and the pharmaceutical industry. The Subcommittee’s mission stated in the law was to ensure that prioritization of agents for children with cancer, for the purposes of clinical trials, was consensus driven. The Subcommittee is advisory, with the FDA controlling its agenda, outcomes, and implementation of recommendations. The Pediatric Incentive While the FDA was forming and revising the 1994 and 1998 Pediatric Rules, Congress was working on a complementary strategy of offering companies an economic incentive to conduct studies in children in response to formal requests from the FDA. The incentive is a 6-month extension of marketing exclusivity for the company’s products containing the active ingredient studied (the “active moiety”), and applies both to approved and investigational drugs, but not to biologics. The pediatric exclusivity provision was written into Section 111 of the Food and Drug Administration Modernization Act (FDAMA) of 1997. Depending on the drug, studies requested might focus on one or more of the following: safety, efficacy, pharmacokinetics, pharmacodynamics, or other topics.

OCR for page 1
Making Better Drugs for Children with Cancer FDAMA set out tasks to establish the exclusivity provision. Among the first was publication by the Secretary of Health and Human Services of a list of then-approved drugs for which additional pediatric information was likely to produce health benefits for children. The list, which is updated annually, consists of all drugs approved for use in adults that could have indications for use in children. A company may choose to carry out studies, mutually agreed upon with the FDA, on any of its drugs on the list. The pediatric incentive has been extended through 2007 by incorporation into the BPCA. The Pediatric Incentive has had a modest effect on pediatric cancer drugs. Over the life span of the pediatric incentive, in its various forms, the FDA reports issuing 31 written requests for studies of oncology products (U.S. Food and Drug Administration, 2005). Sixteen are readily identifiable as cancer related in the enumerated list of products (Table 5). Ten of these requests resulted in extensions of exclusivity, and in six of these cases, labels were changed, most adding information about dosing and side effects. The incentive has not had an effect of moving forward the initiation of pediatric clinical trials. A company that conducts requested studies any time during the patent life of a drug is eligible for the 6 months of exclusivity, so there is no incentive for them to do so early in the life of the drug.

OCR for page 1
Making Better Drugs for Children with Cancer TABLE 5 Approved Oncology Drugs for Which the FDA Has Issued Written Requests for Studies Under the Best Pharmaceuticals for Children Acta Drug Name Pediatric Exclusivity Granted by FDAb Pediatric Label Changesc Summary of Medical and Clinical Pharmacology Reviews Postedd Busulfan x x   Carboplatin x   x Cytarabine       Epirubicin       Fentanyl x x x Fludarabine x x x Gemcitabine x     Gemtuzumab       Hydroxyurea       Imatinib       Irinotecan x x x Oxaliplatin       Tamoxifene x x   Temozolomide x x x Topotecan x   x Vinorelbine x     ahttp://www.fda.gov/cder/pediatric/wrlist.htm. Last updated February 7, 2005. The text introducing this table states that these are requests made as of June 1998, so the coverage is not entirely clear. bhttp://www.fda.gov/cder/pediatric/exgrant.htm. Last updated February 14, 2005. chttp://www.fda.gov/cder/pediatric/labelchange.htm. Last updated January 7, 2005. dhttp://www.fda.gov/cder/pediatric/Summaryreview.htm. Summaries are posted for reports received in response to BPCA requests, but presumably not those submitted earlier, under the pre-BPCA incentive rules. Last updated January 28, 2005. eTamoxifen has no pediatric cancer indication. It was studied in female patients aged 2 to 10 years with McCune-Albright Syndrome and precocious puberty.

OCR for page 1
Making Better Drugs for Children with Cancer RECOMMENDATIONS A new public–private partnership, involving government, industry, academic and other research institutions, advocacy groups, philanthropies, and others, should be formed to lead pediatric cancer drug discovery and development. Advocacy groups, government, academic institutions, and the pharmaceutical and biotechnology industries should collaborate to establish a not-for-profit “public–private partnership” to fund and direct drug development for agents uniquely targeting pediatric cancers. Arrangements could be made, for example, for industry to relinquish intellectual property rights (or otherwise reduce barriers) associated with pediatric use of candidate compounds identified by screening their libraries against pediatric targets (while retaining rights for any other uses); for academic researchers and their institutions to act similarly regarding potential drug targets (also a contribution of intellectual property); for NCI to make available its drug development and clinical trial resources on a preferential basis; and for advocacy groups to contribute input from the patient community, fundraising, and leadership. A first step would be for each partner (industry, government, academia, research institutions, advocacy) to identify and support an individual to work for a period of time on a feasibility study, business plan (in consultation with professional business development experts), and plans for pilot projects, which would be seen as a means to prove the viability of the concept. The components of R&D for pediatric cancer drugs are, in many ways, better developed than for most other rare conditions, and money is already flowing to many of them. This entity could take advantage of those resources in a more focused way than currently occurs for pediatric cancer drugs. The National Cancer Institute should assume responsibility as the developer of last resort for agents that show promise only in children if companies decide not to proceed with full-scale development. Because of the small market potential, companies may decide against completing the development of agents that show promise largely for treating childhood cancers. If the public–private partnership described in the first recommendation is formed, NCI could carry out this work in conjunction with that organization. In addition, NCI resources should be made available to assist in development and production of pediatric dosage forms. NCI already has the programs in place to carry out these tasks, but they have not been focused specifically on pediatric drugs.

OCR for page 1
Making Better Drugs for Children with Cancer The pharmaceutical industry, National Cancer Institute, and Food and Drug Administration should act to reduce the delay in beginning pediatric clinical studies of agents in development for adult cancers. NCI, pediatric oncology researchers, and other interested parties should continue to explore ways to test compounds in development for adult cancers in preclinical and early clinical trials as soon, and as quickly, as possible. Specifically: The evolving NCI preclinical testing program should be fully supported and eventually expanded because it will play a key role in providing the evidence needed to begin early-stage clinical trials of new agents in children. NCI should encourage and facilitate international consortia of pediatric cancer centers and clinical trials groups to provide a larger base of available patients for testing new anticancer agents. The FDA should (with assistance from NCI and others) assemble a packet of information on resources for preclinical pediatric drug testing, which would be provided to every company submitting an IND for a new molecular entity for cancer.